--- license: cc-by-nc-4.0 base_model: MCG-NJU/videomae-small-finetuned-kinetics tags: - generated_from_trainer metrics: - accuracy - matthews_correlation model-index: - name: videomae-small-finetuned-kinetics-finetuned-SNchunks-5c-a40 results: [] --- # videomae-small-finetuned-kinetics-finetuned-SNchunks-5c-a40 This model is a fine-tuned version of [MCG-NJU/videomae-small-finetuned-kinetics](https://huggingface.co/MCG-NJU/videomae-small-finetuned-kinetics) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.7599 - Accuracy: 0.7159 - Balanced Accuracy: 0.7157 - Matthews Correlation: 0.6515 - Confusion Matrix: [[1135 54 69 73 41] [ 333 828 92 50 68] [ 161 23 1008 165 13] [ 306 34 292 705 27] [ 102 17 16 9 1226]] - 0 Ball out of play: {'precision': 0.5571919489445263, 'recall': 0.827259475218659, 'f1-score': 0.665884423584629, 'support': 1372.0} - Precision 0: 0.5572 - Recall 0: 0.8273 - F1-score 0: 0.6659 - Support 0: 1372.0 - 1 Foul: {'precision': 0.8661087866108786, 'recall': 0.6039387308533917, 'f1-score': 0.7116458960034379, 'support': 1371.0} - Precision 1: 0.8661 - Recall 1: 0.6039 - F1-score 1: 0.7116 - Support 1: 1371.0 - 2 Goal: {'precision': 0.6824644549763034, 'recall': 0.7357664233576642, 'f1-score': 0.7081138040042149, 'support': 1370.0} - Precision 2: 0.6825 - Recall 2: 0.7358 - F1-score 2: 0.7081 - Support 2: 1370.0 - 3 Shots: {'precision': 0.7035928143712575, 'recall': 0.5168621700879765, 'f1-score': 0.5959425190194421, 'support': 1364.0} - Precision 3: 0.7036 - Recall 3: 0.5169 - F1-score 3: 0.5959 - Support 3: 1364.0 - 4 Throw-in: {'precision': 0.8916363636363637, 'recall': 0.8948905109489051, 'f1-score': 0.8932604735883425, 'support': 1370.0} - Precision 4: 0.8916 - Recall 4: 0.8949 - F1-score 4: 0.8933 - Support 4: 1370.0 - Precision Macro avg: 0.7402 - Recall Macro avg: 0.7157 - F1-score Macro avg: 0.7150 - Support Macro avg: 6847.0 - Precision Weighted avg: 0.7402 - Recall Weighted avg: 0.7159 - F1-score Weighted avg: 0.7151 - Support Weighted avg: 6847.0 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 16 - eval_batch_size: 16 - seed: 42 - distributed_type: multi-GPU - num_devices: 2 - gradient_accumulation_steps: 2 - total_train_batch_size: 64 - total_eval_batch_size: 32 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_ratio: 0.1 - num_epochs: 10 ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | Balanced Accuracy | Matthews Correlation | Confusion Matrix | 0 Ball out of play | Precision 0 | Recall 0 | F1-score 0 | Support 0 | 1 Foul | Precision 1 | Recall 1 | F1-score 1 | Support 1 | 2 Goal | Precision 2 | Recall 2 | F1-score 2 | Support 2 | 3 Shots | Precision 3 | Recall 3 | F1-score 3 | Support 3 | 4 Throw-in | Precision 4 | Recall 4 | F1-score 4 | Support 4 | Precision Macro avg | Recall Macro avg | F1-score Macro avg | Support Macro avg | Precision Weighted avg | Recall Weighted avg | F1-score Weighted avg | Support Weighted avg | |:-------------:|:-----:|:----:|:---------------:|:--------:|:-----------------:|:--------------------:|:--------------------------------------------------------------------------------------------------------------------------------------------:|:------------------------------------------------------------------------------------------------------------------:|:-----------:|:--------:|:----------:|:---------:|:-------------------------------------------------------------------------------------------------------------------:|:-----------:|:--------:|:----------:|:---------:|:------------------------------------------------------------------------------------------------------------------:|:-----------:|:--------:|:----------:|:---------:|:--------------------------------------------------------------------------------------------------------------------:|:-----------:|:--------:|:----------:|:---------:|:------------------------------------------------------------------------------------------------------------------:|:-----------:|:--------:|:----------:|:---------:|:-------------------:|:----------------:|:------------------:|:-----------------:|:----------------------:|:-------------------:|:---------------------:|:--------------------:| | 0.7825 | 1.0 | 428 | 0.9234 | 0.6207 | 0.6205 | 0.5371 | [[ 961 43 100 80 188] [ 461 597 105 35 173] [ 195 22 953 167 33] [ 454 33 354 464 59] [ 65 7 19 4 1275]] | {'precision': 0.4499063670411985, 'recall': 0.7004373177842566, 'f1-score': 0.5478905359179019, 'support': 1372.0} | 0.4499 | 0.7004 | 0.5479 | 1372.0 | {'precision': 0.8504273504273504, 'recall': 0.43544857768052514, 'f1-score': 0.5759768451519537, 'support': 1371.0} | 0.8504 | 0.4354 | 0.5760 | 1371.0 | {'precision': 0.6224689745264533, 'recall': 0.6956204379562044, 'f1-score': 0.6570148224750086, 'support': 1370.0} | 0.6225 | 0.6956 | 0.6570 | 1370.0 | {'precision': 0.6186666666666667, 'recall': 0.34017595307917886, 'f1-score': 0.4389782403027436, 'support': 1364.0} | 0.6187 | 0.3402 | 0.4390 | 1364.0 | {'precision': 0.7378472222222222, 'recall': 0.9306569343065694, 'f1-score': 0.8231116849580373, 'support': 1370.0} | 0.7378 | 0.9307 | 0.8231 | 1370.0 | 0.6559 | 0.6205 | 0.6086 | 6847.0 | 0.6559 | 0.6207 | 0.6087 | 6847.0 | | 0.8655 | 2.0 | 856 | 0.8769 | 0.6648 | 0.6646 | 0.5874 | [[1007 78 82 82 123] [ 328 794 78 54 117] [ 162 36 972 182 18] [ 398 50 313 536 67] [ 74 15 30 8 1243]] | {'precision': 0.5114271203656678, 'recall': 0.7339650145772595, 'f1-score': 0.6028135288835678, 'support': 1372.0} | 0.5114 | 0.7340 | 0.6028 | 1372.0 | {'precision': 0.816032887975334, 'recall': 0.5791393143690736, 'f1-score': 0.6774744027303755, 'support': 1371.0} | 0.8160 | 0.5791 | 0.6775 | 1371.0 | {'precision': 0.6589830508474577, 'recall': 0.7094890510948905, 'f1-score': 0.6833040421792619, 'support': 1370.0} | 0.6590 | 0.7095 | 0.6833 | 1370.0 | {'precision': 0.6218097447795824, 'recall': 0.39296187683284456, 'f1-score': 0.48158131176999097, 'support': 1364.0} | 0.6218 | 0.3930 | 0.4816 | 1364.0 | {'precision': 0.7927295918367347, 'recall': 0.9072992700729927, 'f1-score': 0.8461538461538461, 'support': 1370.0} | 0.7927 | 0.9073 | 0.8462 | 1370.0 | 0.6802 | 0.6646 | 0.6583 | 6847.0 | 0.6802 | 0.6648 | 0.6584 | 6847.0 | | 0.8065 | 3.0 | 1284 | 0.7639 | 0.7037 | 0.7035 | 0.6356 | [[1046 89 119 82 36] [ 271 906 104 47 43] [ 106 21 1116 126 1] [ 266 35 408 646 9] [ 141 51 60 14 1104]] | {'precision': 0.571584699453552, 'recall': 0.7623906705539358, 'f1-score': 0.6533416614615865, 'support': 1372.0} | 0.5716 | 0.7624 | 0.6533 | 1372.0 | {'precision': 0.822141560798548, 'recall': 0.6608315098468271, 'f1-score': 0.7327133036797413, 'support': 1371.0} | 0.8221 | 0.6608 | 0.7327 | 1371.0 | {'precision': 0.6175982291090205, 'recall': 0.8145985401459854, 'f1-score': 0.7025495750708216, 'support': 1370.0} | 0.6176 | 0.8146 | 0.7025 | 1370.0 | {'precision': 0.7060109289617487, 'recall': 0.4736070381231672, 'f1-score': 0.566915313734094, 'support': 1364.0} | 0.7060 | 0.4736 | 0.5669 | 1364.0 | {'precision': 0.9253981559094719, 'recall': 0.8058394160583942, 'f1-score': 0.8614904408895825, 'support': 1370.0} | 0.9254 | 0.8058 | 0.8615 | 1370.0 | 0.7285 | 0.7035 | 0.7034 | 6847.0 | 0.7285 | 0.7037 | 0.7035 | 6847.0 | | 0.6598 | 4.0 | 1712 | 0.7694 | 0.6994 | 0.6992 | 0.6319 | [[1106 42 82 80 62] [ 379 735 117 60 80] [ 133 17 1053 159 8] [ 293 28 340 671 32] [ 98 16 21 11 1224]] | {'precision': 0.5505226480836237, 'recall': 0.8061224489795918, 'f1-score': 0.6542443064182195, 'support': 1372.0} | 0.5505 | 0.8061 | 0.6542 | 1372.0 | {'precision': 0.8770883054892601, 'recall': 0.5361050328227571, 'f1-score': 0.6654594839293798, 'support': 1371.0} | 0.8771 | 0.5361 | 0.6655 | 1371.0 | {'precision': 0.652820830750155, 'recall': 0.7686131386861313, 'f1-score': 0.7060006704659737, 'support': 1370.0} | 0.6528 | 0.7686 | 0.7060 | 1370.0 | {'precision': 0.6839959225280327, 'recall': 0.49193548387096775, 'f1-score': 0.5722814498933901, 'support': 1364.0} | 0.6840 | 0.4919 | 0.5723 | 1364.0 | {'precision': 0.8705547652916074, 'recall': 0.8934306569343066, 'f1-score': 0.8818443804034583, 'support': 1370.0} | 0.8706 | 0.8934 | 0.8818 | 1370.0 | 0.7270 | 0.6992 | 0.6960 | 6847.0 | 0.7270 | 0.6994 | 0.6961 | 6847.0 | | 0.5968 | 5.0 | 2140 | 0.7820 | 0.6991 | 0.6989 | 0.6335 | [[1140 50 77 59 46] [ 360 834 85 32 60] [ 186 26 1007 140 11] [ 384 56 293 593 38] [ 129 19 6 3 1213]] | {'precision': 0.5184174624829468, 'recall': 0.8309037900874635, 'f1-score': 0.6384766171940633, 'support': 1372.0} | 0.5184 | 0.8309 | 0.6385 | 1372.0 | {'precision': 0.8467005076142132, 'recall': 0.6083150984682714, 'f1-score': 0.7079796264855689, 'support': 1371.0} | 0.8467 | 0.6083 | 0.7080 | 1371.0 | {'precision': 0.6859673024523161, 'recall': 0.7350364963503649, 'f1-score': 0.7096546863988723, 'support': 1370.0} | 0.6860 | 0.7350 | 0.7097 | 1370.0 | {'precision': 0.717049576783555, 'recall': 0.4347507331378299, 'f1-score': 0.5413053400273847, 'support': 1364.0} | 0.7170 | 0.4348 | 0.5413 | 1364.0 | {'precision': 0.8866959064327485, 'recall': 0.8854014598540146, 'f1-score': 0.8860482103725348, 'support': 1370.0} | 0.8867 | 0.8854 | 0.8860 | 1370.0 | 0.7310 | 0.6989 | 0.6967 | 6847.0 | 0.7309 | 0.6991 | 0.6968 | 6847.0 | | 0.5675 | 6.0 | 2568 | 0.7603 | 0.7159 | 0.7157 | 0.6515 | [[1135 54 69 73 41] [ 333 828 92 50 68] [ 161 23 1008 165 13] [ 306 34 292 705 27] [ 102 17 16 9 1226]] | {'precision': 0.5571919489445263, 'recall': 0.827259475218659, 'f1-score': 0.665884423584629, 'support': 1372.0} | 0.5572 | 0.8273 | 0.6659 | 1372.0 | {'precision': 0.8661087866108786, 'recall': 0.6039387308533917, 'f1-score': 0.7116458960034379, 'support': 1371.0} | 0.8661 | 0.6039 | 0.7116 | 1371.0 | {'precision': 0.6824644549763034, 'recall': 0.7357664233576642, 'f1-score': 0.7081138040042149, 'support': 1370.0} | 0.6825 | 0.7358 | 0.7081 | 1370.0 | {'precision': 0.7035928143712575, 'recall': 0.5168621700879765, 'f1-score': 0.5959425190194421, 'support': 1364.0} | 0.7036 | 0.5169 | 0.5959 | 1364.0 | {'precision': 0.8916363636363637, 'recall': 0.8948905109489051, 'f1-score': 0.8932604735883425, 'support': 1370.0} | 0.8916 | 0.8949 | 0.8933 | 1370.0 | 0.7402 | 0.7157 | 0.7150 | 6847.0 | 0.7402 | 0.7159 | 0.7151 | 6847.0 | | 0.4824 | 7.0 | 2996 | 0.8064 | 0.6958 | 0.6956 | 0.6308 | [[1178 37 62 69 26] [ 396 787 80 57 51] [ 188 14 993 172 3] [ 378 32 287 650 17] [ 173 16 17 8 1156]] | {'precision': 0.5092952875054042, 'recall': 0.858600583090379, 'f1-score': 0.639348710990502, 'support': 1372.0} | 0.5093 | 0.8586 | 0.6393 | 1372.0 | {'precision': 0.8882618510158014, 'recall': 0.574033552151714, 'f1-score': 0.6973859105006646, 'support': 1371.0} | 0.8883 | 0.5740 | 0.6974 | 1371.0 | {'precision': 0.6900625434329395, 'recall': 0.7248175182481752, 'f1-score': 0.7070131719473122, 'support': 1370.0} | 0.6901 | 0.7248 | 0.7070 | 1370.0 | {'precision': 0.6799163179916318, 'recall': 0.47653958944281527, 'f1-score': 0.560344827586207, 'support': 1364.0} | 0.6799 | 0.4765 | 0.5603 | 1364.0 | {'precision': 0.922585794094174, 'recall': 0.8437956204379562, 'f1-score': 0.881433473122379, 'support': 1370.0} | 0.9226 | 0.8438 | 0.8814 | 1370.0 | 0.7380 | 0.6956 | 0.6971 | 6847.0 | 0.7380 | 0.6958 | 0.6972 | 6847.0 | | 0.6574 | 8.0 | 3424 | 0.7998 | 0.7035 | 0.7033 | 0.6385 | [[1141 55 85 65 26] [ 341 827 113 50 40] [ 150 19 1084 113 4] [ 321 47 353 624 19] [ 166 32 25 6 1141]] | {'precision': 0.5384615384615384, 'recall': 0.8316326530612245, 'f1-score': 0.6536808937267259, 'support': 1372.0} | 0.5385 | 0.8316 | 0.6537 | 1372.0 | {'precision': 0.8438775510204082, 'recall': 0.6032093362509118, 'f1-score': 0.7035304125903871, 'support': 1371.0} | 0.8439 | 0.6032 | 0.7035 | 1371.0 | {'precision': 0.653012048192771, 'recall': 0.7912408759124088, 'f1-score': 0.7155115511551154, 'support': 1370.0} | 0.6530 | 0.7912 | 0.7155 | 1370.0 | {'precision': 0.7272727272727273, 'recall': 0.4574780058651026, 'f1-score': 0.5616561656165616, 'support': 1364.0} | 0.7273 | 0.4575 | 0.5617 | 1364.0 | {'precision': 0.9276422764227642, 'recall': 0.8328467153284671, 'f1-score': 0.8776923076923077, 'support': 1370.0} | 0.9276 | 0.8328 | 0.8777 | 1370.0 | 0.7381 | 0.7033 | 0.7024 | 6847.0 | 0.7380 | 0.7035 | 0.7025 | 6847.0 | | 0.4709 | 9.0 | 3852 | 0.8032 | 0.7024 | 0.7021 | 0.6373 | [[1161 47 70 68 26] [ 365 794 98 62 52] [ 177 16 1019 155 3] [ 353 39 297 654 21] [ 149 19 16 5 1181]] | {'precision': 0.5265306122448979, 'recall': 0.8462099125364432, 'f1-score': 0.6491473301649426, 'support': 1372.0} | 0.5265 | 0.8462 | 0.6491 | 1372.0 | {'precision': 0.8677595628415301, 'recall': 0.5791393143690736, 'f1-score': 0.6946631671041119, 'support': 1371.0} | 0.8678 | 0.5791 | 0.6947 | 1371.0 | {'precision': 0.6793333333333333, 'recall': 0.7437956204379562, 'f1-score': 0.7101045296167248, 'support': 1370.0} | 0.6793 | 0.7438 | 0.7101 | 1370.0 | {'precision': 0.6927966101694916, 'recall': 0.47947214076246336, 'f1-score': 0.5667244367417678, 'support': 1364.0} | 0.6928 | 0.4795 | 0.5667 | 1364.0 | {'precision': 0.9204988308651598, 'recall': 0.862043795620438, 'f1-score': 0.8903128533735394, 'support': 1370.0} | 0.9205 | 0.8620 | 0.8903 | 1370.0 | 0.7374 | 0.7021 | 0.7022 | 6847.0 | 0.7374 | 0.7024 | 0.7023 | 6847.0 | | 0.3689 | 10.0 | 4280 | 0.8093 | 0.7082 | 0.7079 | 0.6447 | [[1160 58 65 58 31] [ 343 852 86 40 50] [ 191 23 1015 136 5] [ 383 52 284 624 21] [ 130 24 13 5 1198]] | {'precision': 0.5256003624830086, 'recall': 0.8454810495626822, 'f1-score': 0.648225761385862, 'support': 1372.0} | 0.5256 | 0.8455 | 0.6482 | 1372.0 | {'precision': 0.844400396432111, 'recall': 0.6214442013129103, 'f1-score': 0.7159663865546219, 'support': 1371.0} | 0.8444 | 0.6214 | 0.7160 | 1371.0 | {'precision': 0.69377990430622, 'recall': 0.7408759124087592, 'f1-score': 0.7165548888104484, 'support': 1370.0} | 0.6938 | 0.7409 | 0.7166 | 1370.0 | {'precision': 0.7230590961761297, 'recall': 0.4574780058651026, 'f1-score': 0.5603951504265828, 'support': 1364.0} | 0.7231 | 0.4575 | 0.5604 | 1364.0 | {'precision': 0.918007662835249, 'recall': 0.8744525547445255, 'f1-score': 0.8957009345794392, 'support': 1370.0} | 0.9180 | 0.8745 | 0.8957 | 1370.0 | 0.7410 | 0.7079 | 0.7074 | 6847.0 | 0.7409 | 0.7082 | 0.7075 | 6847.0 | ### Framework versions - Transformers 4.31.0 - Pytorch 2.0.1+git8bfa463 - Datasets 2.13.1 - Tokenizers 0.13.3